JP2007080014A - Image evaluation device and method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To calculate an integration evaluation value obtained by easily and properly integrating evaluation results about a plurality of evaluation items, when selecting at least one image from among a plurality of images by using the plurality of evaluation items. <P>SOLUTION: Evaluation parts 30, 32, 34 respectively calculate a plurality of feature amounts indicating the evaluation results about the plurality of evaluation items, and respectively statistically calculate the probability that indicates whether the image is to be selected, on the basis of the plurality of calculated characteristic amounts. An integration evaluation part 36 calculates the integration evaluation value that is an integrated evaluation value of the image, on the basis of a plurality of evaluation values. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image evaluation apparatus and method for evaluating an image in order to select, for example, an image recommended for printing from a plurality of images, and a program for causing a computer to execute the image evaluation method.

  In recent years, the spread of digital cameras and the increase in the capacity of recording media for storing images have made it possible for users to hold many images in one medium. . On the other hand, there is a trouble of selecting an image to be processed such as printing from a large number of images, and in order to select an image efficiently, the user finally determines that the image is to be printed. There is a demand for a function for narrowing down an image under a certain condition and a function for selecting an image suitable for printing according to the user's preference.

  For example, Patent Document 1 discloses a method of evaluating an image based on any one of image brightness, an acceleration sensor, and AF evaluation, and automatically omitting a photograph inappropriate for printing using the evaluation result. Yes.

  Japanese Patent Application Laid-Open No. 2004-228620 determines a face orientation of a person included in an image, calculates an image evaluation value based on the determination result, and selects an image from a plurality of images based on the calculated evaluation value. Is disclosed.

Further, in Patent Document 3, an image is evaluated using an evaluation result obtained by integrating a plurality of evaluation items such as a ratio of a face in an image, a degree of opening of an eye, a face direction, focus, blur, and brightness. A method for selecting a high-quality image from a plurality of images is disclosed.
JP 2002-10179 A JP 2004-361989 A JP 2002-358522 A

  In the method disclosed in Patent Document 3 described above, evaluation is performed for each of a plurality of evaluation items, and image evaluation is performed by combining evaluations for each item, but all evaluation items are evaluated with high performance. Therefore, in order to perform comprehensive evaluation, it is necessary to determine a method for calculating a comprehensive evaluation value after grasping evaluation performance for each evaluation item. For example, if the performance of the method for evaluating the brightness is low or the performance of the method for evaluating the orientation of the face is high, the brightness evaluation result is used when determining the overall evaluation value. It is necessary to reduce the degree of consideration of the image quality or to increase the degree of consideration of the evaluation result of the face orientation. For this reason, the design of the method for calculating the comprehensive evaluation value becomes complicated, and the processing for calculating the comprehensive evaluation value also becomes complicated.

  The present invention has been made in view of the above circumstances, and an object thereof is to be able to calculate a comprehensive evaluation value obtained by simply and appropriately integrating evaluation results for a plurality of evaluation items.

An image evaluation apparatus according to the present invention is an image evaluation apparatus that evaluates the image in order to select at least one image from a plurality of images,
A plurality of feature amounts representing evaluation results for a plurality of evaluation items are calculated for each of the plurality of images to be processed, and the image is selected based on the calculated plurality of feature amounts. A plurality of evaluation means for statistically calculating a probability representing whether each of the plurality of evaluation items and outputting an evaluation value based on the calculated plurality of probabilities;
Comprehensive evaluation means for calculating a comprehensive evaluation value, which is a comprehensive evaluation value of the image based on the plurality of evaluation values, is provided.

  The “evaluation item” means an item that can represent the evaluation result as a feature amount. Specifically, the average brightness of the image, the number of faces included in the image, the position of the face included in the image, Face evaluation, face size, facial expression, importance of event at the time of image acquisition, number of similar images, and the like can be used as evaluation items.

  The “probability representing whether an image is selected” is an index representing a ratio of an image that is a certain feature amount selected from a plurality of images having the same feature amount with respect to a certain evaluation item. For example, when there are 100 images with a feature value of 150 as the brightness evaluation item, and the number of images selected from the 100 images is statistically 80, “represents whether or not an image is selected”. The “probability” is 0.8.

  The “evaluation value based on probability” may be the probability value itself, a value expressing the probability as a percentage, a value obtained by subtracting a value of 0.5 from the probability, or a value indicating a probability expressed as a percentage of 50. A value obtained by subtracting can be used. For example, when the “probability indicating whether an image is selected” is 0.8, the “evaluation value based on probability” is 0.8, 80%, 0.3, 30%, or the like.

  Note that the image evaluation apparatus according to the present invention may further include image selection means for selecting the at least one image from the plurality of images based on the comprehensive evaluation value.

In this case, it is further provided with a designation receiving means for receiving designation of an image selection rate from the plurality of images,
The image selection means may be means for selecting, from the plurality of images, a number of images corresponding to the designated selection rate based on the comprehensive evaluation value.

  In this case, display control means for displaying the selected at least one image on the display means may be further provided.

  In this case, the display control means may be means for displaying a list of the plurality of images on the display means together with the at least one selected image.

  In the image evaluation apparatus according to the present invention, the evaluation unit may be a unit that calculates the probability using a look-up table or a function that is statistically defined with respect to a relationship between the feature quantity and the probability.

  In this case, the evaluation means may be means for updating the lookup table or the function based on the actual selection result of the image.

  In the image evaluation apparatus according to the present invention, the comprehensive evaluation unit calculates the comprehensive evaluation value based on at least one of a sum of the plurality of evaluation values and a maximum value and a minimum value of the plurality of evaluation values. It may be a means.

Another image evaluation apparatus according to the present invention is an image evaluation apparatus for evaluating the image in order to select at least one image from a plurality of images,
A feature amount is calculated for an image to be processed among the plurality of images, and a probability indicating whether or not the image is selected is statistically calculated based on the calculated feature amount, and the calculated Evaluation means for outputting evaluation values based on a plurality of probabilities is provided.

An image evaluation method according to the present invention is an image evaluation method for evaluating the image in order to select at least one image from a plurality of images,
For each of the plurality of images to be processed, a plurality of feature amounts each representing an evaluation result for a plurality of evaluation items are calculated,
Based on the calculated plurality of feature amounts, a probability representing whether or not the image is selected is statistically calculated for each of the plurality of evaluation items,
Each of the evaluation values based on the calculated plurality of probabilities is output,
A comprehensive evaluation value that is a comprehensive evaluation value of the image is calculated based on the plurality of evaluation values.

  The image evaluation method according to the present invention may be provided as a program for causing a computer to execute the image evaluation method.

  According to the present invention, for a processing target image, a plurality of feature amounts representing evaluation results for a plurality of evaluation items are respectively calculated, and whether or not an image is selected based on the calculated plurality of feature amounts. Is statistically calculated for each of the plurality of evaluation items. Then, evaluation values based on the plurality of calculated probabilities are output, and a comprehensive evaluation value that is a comprehensive evaluation value of the image is calculated based on the plurality of output evaluation values.

  Here, when the performance of the evaluation item evaluation method, that is, the method for calculating the feature amount is high, the evaluation level can be clearly defined, so that selection and non-selection of the image are clearly separated. As a result, the probability is divided into a low value and a high value on both sides of 0.5. On the other hand, if the performance of the technique for calculating the feature amount is low, it is not possible to clearly determine the evaluation level, so it is not possible to clearly separate the selection and non-selection of the image. Takes a value in the vicinity of 0.5. For this reason, the evaluation value based on the probability becomes a value farther from 0.5 as the performance of the evaluation method of the evaluation item becomes higher, and becomes closer to 0.5 as the performance of the evaluation method of the evaluation item becomes lower. Therefore, by calculating the overall evaluation value based on the plurality of calculated evaluation values, the overall evaluation value of the image can be easily and appropriately calculated without considering the performance of the evaluation item evaluation method.

  Moreover, an image can be appropriately selected by selecting at least one image from a plurality of images based on the comprehensive evaluation value.

  Also, designation of image selection rates from a plurality of images is accepted, and a desired number of images are selected from a plurality of images according to the number of selections specified based on the overall evaluation value. A number of images corresponding to the ratio can be selected.

  Further, if at least one selected image is displayed, it is possible to immediately recognize what image has been selected.

  Further, if a list of a plurality of images is displayed together with at least one selected image, the selected image and the plurality of images can be easily compared.

  In addition, the probability can be quickly calculated by calculating the probability using a look-up table or a function that is statistically defined regarding the relationship between the feature quantity and the probability.

  In addition, by updating the look-up table or function based on the actual selection result of the image, it is possible to evaluate the image reflecting the user's preference for selecting the image using the present invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram showing the configuration of an image evaluation apparatus according to the first embodiment of the present invention. As shown in FIG. 1, the image evaluation apparatus 1 according to the present embodiment performs various controls such as recording control and display control of image data, and controls each part of the apparatus 1 and operates the CPU 12. A system memory 14 composed of a program, viewer software for viewing images, a ROM in which various constants are recorded, and a RAM serving as a work area when the CPU 12 executes processing, and various instructions to the apparatus 1 An input unit 16 including a keyboard and a mouse for performing the display and a display unit 18 including a liquid crystal monitor for performing various displays.

  The image evaluation apparatus 1 also includes a card slot 20 for reading image data from the memory card 2 on which the image data is recorded and recording image data on the memory card 2, and a technique represented by JPEG for the image data. A compression / decompression unit 22 that compresses or decompresses compressed image data; a hard disk 24 that records image data and various programs executed by the CPU 12; and a memory control that controls the system memory 14, the card slot 20, and the hard disk 24 Unit 26 and a display control unit 28 for controlling the display of the display unit 18.

  Further, the image evaluation apparatus 1 is based on a plurality (three in this case) of the evaluation units 30, 32, and 34 that output evaluation values for a plurality of evaluation items, and the evaluation values output by the evaluation units 30, 32, and 34. A comprehensive evaluation unit 36 for calculating a comprehensive evaluation value of the image, and a plurality of images (for example, all images recorded in the memory card 2) to be processed based on the comprehensive evaluation value 36 calculated by the comprehensive evaluation unit 36. And an image selection unit 38 for selecting an image recommended for printing.

  Each of the evaluation units 30, 32, and 34 calculates a plurality of feature amounts representing evaluation results for different evaluation items for the image. Here, in the present embodiment, the evaluation unit 30 evaluates the brightness of the image, the evaluation unit 32 evaluates the number of faces included in the image, and the evaluation unit 34 evaluates the position of the face included in the image. To do. It is assumed that images are applied to all images recorded on the memory card 2.

  Specifically, the evaluation unit 30 calculates an average luminance value of the image as a feature amount. Here, the average luminance value is a value obtained by dividing the sum of the pixel values of all the pixels included in the image by the total number of pixels. The evaluation unit 32 detects a human face from the image, and calculates the feature amount by counting the number of detected faces. In addition, the evaluation unit 34 detects a human face from the image, and calculates, for example, a coordinate value on the image of the position of the nose included in the detected face as a feature amount.

  Subsequently, the evaluation units 30, 32, and 34 statistically calculate a probability indicating whether or not an image is selected based on the calculated feature amount. Specifically, the probability is calculated using a look-up table that statistically defines the relationship between the feature quantity and the probability indicating whether or not an image is selected.

  2A to 2C are diagrams showing look-up tables used by the evaluation units 30, 32, and 34, respectively. First, the lookup table LUT1 shown in FIG. 2A prints in the past about each of the average luminance values (0 to 255) when the image is 8 bits and a plurality of images having various average luminance values. Probabilities Pk0 to Pk255 selected as appropriate are associated with each other. Note that the probability here refers to a percentage representing how many of the images having a certain average luminance value are selected for printing.

  In the lookup table LUT2 shown in FIG. 2B, each of the number of faces 0 to 10 or more included in the image and the probabilities Ps0 to Ps11 selected to print in the past for each number of faces are included. It is associated. The probability here is a percentage indicating the number of images selected for printing when there are a plurality of images of a certain number of faces.

  The look-up table LUT3 shown in FIG. 2C is obtained by dividing each of the positions on the face image area included in the image and the position of each face when the image is divided into 20 areas A1 to A20 of 5 × 4. For the image at the position, the probabilities Pp0 to Pp20 selected to be printed in the past are associated with each other. Note that the probability here indicates a percentage indicating how many of the images at a certain face position are selected for printing.

  Then, the evaluation units 30, 32, and 34 output values obtained by subtracting 50 from the calculated probabilities to the comprehensive evaluation unit 36 as evaluation values E1 to E3.

  Here, the feature quantity represents the probability as a percentage, but the probability itself may be used as the feature quantity. In this case, the evaluation value is a value obtained by subtracting 0.5 from the calculated probability. Alternatively, the calculated probability value itself or the probability value expressed as a percentage may be output as the evaluation value.

  The look-up tables LUT1 to LUT3 may be stored in the evaluation units 30, 32, and 34, respectively, stored in the hard disk 24, and the look-up stored in the hard disk 24 when the evaluation values E1 to E3 are calculated. The up tables LUT1 to LUT3 may be referred to.

  The comprehensive evaluation unit 36 uses the evaluation values E1 to E3 output from the evaluation units 30, 32, and 34 to calculate a comprehensive evaluation value M0 by the following equation (1).

M0 = ΣEi (i = 1 to 3) (1)
The total evaluation value M0 may be calculated by the following equation (2) in consideration of the maximum value Emax of the evaluation values and the minimum value Emin of the evaluation values.

M0 = αΣEi (i = 1 to 3) + βEmax + (1−α−β) Emin (2)
However, α and β are weight coefficients determined empirically (α + β <1).

  Based on the comprehensive evaluation value calculated by the comprehensive evaluation unit 36, the image selection unit 38 selects a predetermined number of predetermined images from among the images recorded on the memory card 2 as print recommendation images. Specifically, a predetermined upper number of images when images are arranged in descending order of the comprehensive evaluation value is selected as a print recommended image.

  The display control unit 28 generates a list screen including a list of thumbnail images for all images recorded on the memory card 2 and a list of thumbnail images of recommended print images, and the generated list screen is displayed on the display unit 18. indicate.

  FIG. 3 shows a list screen. As shown in FIG. 3, on this list screen 50, a list display area 50A for displaying a list of thumbnail images for all images recorded on the memory card 2 and a recommended print image for displaying a list of recommended print images. Area 50B is included. Here, three images are displayed as recommended print images. In addition, a “recommended” character is displayed in the recommended print image area 50B.

  An image that cannot be displayed in the list display area 50A can be displayed by scrolling the screen using the scroll bar 50C.

  Here, on the list screen shown in FIG. 3, a selection rate bar 51 for designating the selection rate of images from all images recorded on the memory card 2 as shown in FIG. 4 is displayed, and the user selects the selection rate. The bar 51 may be used to specify the image selection rate. The selection rate bar 51 can specify a selection rate by specifying a desired numerical value on the selection rate bar 51 using the input unit 16. In this case, the image selection unit 38 multiplies the designated selection rate by the number of images recorded in the memory card 2 and selects the multiplied number of images as the recommended print image. Specifically, when the number of images recorded on the memory card 2 is 20 and the designated selection rate is 20%, the top four images of the comprehensive evaluation value are selected as recommended print images.

  Next, processing performed in the first embodiment will be described. FIG. 5 is a flowchart showing processing performed in the first embodiment. The CPU 12 starts processing when the user issues an instruction to display a recommended print image from the input unit 16. First, for example, the processing target is set to an image with the file name at the head (i = 1, step ST 1), the evaluation unit 30, 32 and 34 calculate the evaluation values E1 to E3 of the processing target image (step ST2). Then, the comprehensive evaluation unit 36 calculates a comprehensive evaluation value based on the evaluation values E1 to E3 (step ST3). The CPU 12 determines whether or not the processing has been completed for all the images recorded on the memory card 2 (step ST4). If step ST4 is negative, the processing target image is set as the next image (step ST4). ST5), returning to step ST2. If step ST4 is affirmed, the image selection unit 38 selects a predetermined number of images with higher overall evaluation values as recommended print images (step ST6). And the display control part 28 produces | generates a list screen, and displays it on the display part 18 (step ST7), and complete | finishes a process.

  In the list screen, the user can select an image to be ordered by referring to the recommended print image. Here, when the image evaluation apparatus 1 according to the present embodiment is included in a personal computer, the user generates order information by setting the number of prints, the paper size, and the like for the image selected to be printed, and prints this information. It is possible to place a print order for the selected image by recording it in the memory card 2 together with the image to be ordered and bringing it into the photo store, or by sending the order information and the print order image to the photo store via the Internet. In addition, you may make it print out the image selected from the printer connected to the personal computer.

  Further, when the image evaluation apparatus 1 according to the present embodiment is included in a print order receiving apparatus installed at a photo shop, the list screen is displayed on the operation screen of the print order receiving apparatus. The user selects an image to be printed with reference to the recommended print image, and further inputs the number of prints, the paper size, and the like, so that a print order can be placed at a photo shop.

  Here, when the performance of the evaluation item evaluation method, that is, the method for calculating the feature amount is high, the evaluation level can be clearly defined, so that selection and non-selection of the image are clearly separated. As a result, the probability is divided into a low value and a high value on both sides of 50%. On the other hand, if the performance of the technique for calculating the feature value is low, it is not possible to clearly determine the level of evaluation, so it is not possible to clearly separate the selection and non-selection of the image. Takes a value in the vicinity of 50%. Therefore, the evaluation value based on the probability becomes a larger value or a smaller value as the performance of the evaluation item evaluation method is higher, and becomes an intermediate value as the performance of the evaluation item evaluation method is lower.

  In this embodiment, since the comprehensive evaluation value is calculated based on the plurality of calculated evaluation values, the comprehensive evaluation value of the image is calculated easily and appropriately without considering the performance of the evaluation item evaluation method. can do.

  Further, by selecting a print recommendation image from a plurality of images based on the comprehensive evaluation value, it is possible to appropriately select the print recommendation image.

  Further, by using the selection rate bar 51, designation of the ratio of images to be selected from a plurality of images is received, and the number of images corresponding to the designated ratio is selected from the plurality of images based on the comprehensive evaluation value. A desired number of images can be appropriately selected from a plurality of images.

  In the above embodiment, the look-up table for calculating the probability may be updated using the history using the history of the feature amount of the image selected as the recommended print image by the user. Hereinafter, this will be described as a second embodiment.

  FIG. 6 is a schematic block diagram showing the configuration of the image evaluation apparatus according to the second embodiment. In the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted here. The second embodiment is different from the first embodiment in that an operation history management unit 40 is provided.

  The operation history management unit 40 uses the history of the feature amount of the image selected by the user as the recommended print image to update the look-up table for calculating the probability using the history. Hereinafter, the update of the lookup table will be described.

  FIG. 7 is a diagram for explaining the update of the lookup table. Here, updating of the lookup table LUT1 in which the average luminance value is associated with the probability will be described. First, when the user of the image evaluation apparatus 1 according to the present embodiment selects an image to place a print order, the average luminance value of the selected image is calculated, and the ratio of the selected image in the images having the same average luminance value Is calculated as a probability. Each time a print order or print is made, a user lookup table LUT4 is generated as shown in FIG.

  As shown in FIG. 7A, the user lookup table LUT4 associates the average luminance value with the probability selected by the user to print for a plurality of images having various average luminance values. In the user lookup table LUT4, * indicates that no data exists, that is, no image having the average luminance value exists.

  Then, in the generated user lookup table LUT4, the probability corresponding to the average luminance value in which no data exists is calculated by interpolation using the probability of the average luminance value in which data exists, and the user lookup table LUT5 (See FIG. 7B).

  Further, by combining the existing lookup table LUT1 and the user lookup table LUT5, the lookup table LUT1 is updated to generate a new lookup table LUT6 (FIG. 7C). The probability Ri in the new lookup table LUT6 is calculated by the following equation (3).

Ri = αPki + (1−α) Ui (3)
Here, α is a weight coefficient (α <1) determined empirically.

  As described above, by updating the lookup table based on the actual selection result of the image, the image can be evaluated while reflecting the user's preference for selecting the image.

  In the above embodiment, the probability is calculated using the lookup table. However, a function that is statistically defined for the relationship between the feature quantity and the probability is calculated, and the probability is calculated using this function. You may make it do. This function can be obtained by using a least square method or the like for the relationship between the feature quantity and the probability. Also, this function may be updated using the feature amount history of the image selected as the print recommended image by the user.

  Moreover, in the said embodiment, the three evaluation parts 30, 32, and 34 are provided, and in each evaluation part 30, 32, and 34, the brightness of the image, the number of faces included in the image, and the position of the face included in the image However, the evaluation items are not limited to these, and other items such as face orientation, face size, facial expression, importance of event at the time of image acquisition, number of similar images, etc. It may be used as an evaluation item. In this case, an evaluation unit may be provided according to the number of evaluation items.

  In the above embodiment, the list display area 50A and the recommended print image area 50B are provided on the list screen. However, as shown in FIG. 8, only the list display area 53A is provided on the list screen 53, and the recommended print image is displayed. A frame 54 may be added to the selected image.

  In the above embodiment, the recommended print image is selected. However, the image to be selected is not limited to the image recommended for printing.

  Although the apparatus 1 according to the first embodiment of the present invention has been described above, the computer includes the display control unit 28, the evaluation units 30, 32, and 34, the comprehensive evaluation unit 36, the image selection unit 38, and the operation history. A program that functions as means corresponding to the management unit 40 and performs processing as shown in FIG. 5 is also one embodiment of the present invention. A computer-readable recording medium in which such a program is recorded is also one embodiment of the present invention. In these cases, the lookup table may be included in the program or the same recording medium, or may be provided from an external device or a separate medium.

1 is a schematic block diagram showing the configuration of an image evaluation apparatus according to a first embodiment of the present invention. The figure which shows the lookup table which the evaluation section uses Diagram showing list screen Figure showing another example of the list screen The flowchart which shows the process performed in 1st Embodiment. The schematic block diagram which shows the structure of the image evaluation apparatus by the 2nd Embodiment of this invention. Diagram for explaining update of lookup table Figure showing another example of the list screen

Explanation of symbols

1 Image evaluation device 2 Memory card 12 CPU
DESCRIPTION OF SYMBOLS 14 System memory 16 Input part 18 Display part 20 Card slot 22 Compression / decompression part 24 Hard disk 26 Memory control part 28 Display control part 30,32,34 Evaluation part 36 Comprehensive evaluation part 38 Image selection part 40 Operation history management part 50,54 List Screen 51 Selectivity bar

Claims (11)

An image evaluation device for evaluating the image to select at least one image from a plurality of images,
A plurality of feature amounts representing evaluation results for a plurality of evaluation items are calculated for each of the plurality of images to be processed, and the image is selected based on the calculated plurality of feature amounts. A plurality of evaluation means for statistically calculating a probability representing whether each of the plurality of evaluation items and outputting an evaluation value based on the calculated plurality of probabilities;
An image evaluation apparatus comprising: comprehensive evaluation means for calculating a comprehensive evaluation value that is a comprehensive evaluation value of the image based on the plurality of evaluation values.   The image evaluation apparatus according to claim 1, further comprising image selection means for selecting the at least one image from the plurality of images based on the comprehensive evaluation value. Further comprising designation accepting means for accepting designation of an image selection rate from the plurality of images,
The image evaluation apparatus according to claim 2, wherein the image selection unit is a unit that selects, from the plurality of images, a number of images corresponding to the designated selection rate based on the comprehensive evaluation value.   4. The image evaluation apparatus according to claim 2, further comprising display control means for displaying the selected at least one image on a display means.   The image evaluation apparatus according to claim 4, wherein the display control means is a means for displaying a list of the plurality of images on the display means together with the selected at least one image.   6. The evaluation unit according to claim 1, wherein the evaluation unit is a unit that calculates the probability using a look-up table or a function that is statistically defined with respect to a relationship between the feature quantity and the probability. The image evaluation apparatus according to item.   The image evaluation apparatus according to claim 6, wherein the evaluation unit is a unit that updates the lookup table or the function based on an actual selection result of the image.   2. The comprehensive evaluation unit is a unit that calculates the comprehensive evaluation value based on at least one of a sum of the plurality of evaluation values and a maximum value and a minimum value of the plurality of evaluation values. 8. The image evaluation apparatus according to any one of items 1 to 7. An image evaluation device for evaluating the image to select at least one image from a plurality of images,
A feature amount is calculated for an image to be processed among the plurality of images, and a probability indicating whether or not the image is selected is statistically calculated based on the calculated feature amount, and the calculated An image evaluation apparatus comprising evaluation means for outputting evaluation values based on a plurality of probabilities. An image evaluation method for evaluating the image in order to select at least one image from a plurality of images,
For each of the plurality of images to be processed, a plurality of feature amounts each representing an evaluation result for a plurality of evaluation items are calculated,
Based on the calculated plurality of feature amounts, a probability representing whether or not the image is selected is statistically calculated for each of the plurality of evaluation items,
Each of the evaluation values based on the calculated plurality of probabilities is output,
An image evaluation method, comprising: calculating a comprehensive evaluation value that is a comprehensive evaluation value of the image based on the plurality of evaluation values. A program for causing a computer to execute an image evaluation method for evaluating the image in order to select at least one image from a plurality of images,
A procedure for calculating a plurality of feature amounts each representing an evaluation result for a plurality of evaluation items for an image to be processed among the plurality of images,
A step of statistically calculating, for each of the plurality of evaluation items, a probability indicating whether or not the image is selected based on the plurality of calculated feature amounts;
A procedure of outputting evaluation values based on the calculated probabilities,
And a procedure for calculating a comprehensive evaluation value, which is a comprehensive evaluation value of the image, based on the plurality of evaluation values.

Images